There are many factors that contribute to speciation events. Importantly in plants, pollinators can have a large impact on breeding between different individuals. This discrimination of pollinators is driven by a variety of morphological traits in the plant, therefore possibly leading specific pollinators to having a strong plant preference. This knowledge leads to the question: Do pollinators discriminate between subspecies of rabbitbrush? Studying these differences can be important for predicting gene flow and how speciation events may start to occur.

Here is some background information about the study organism to better understand the data presented below.
Rabbitbrush (Ericameria nauseosa) is a native perennial shrub in the Asteraceae family. Rabbitbrush has erect stems and commonly has numerous leaves. When flowers bloom in late summer and fall they are bright yellow and are in clusters. Around 21 species of rabbitbrush have been described within North America and placed in gray or green colored groups (genotypes), both of which can be deciduous or non-deciduous. Within these subspecies ranges are variable with some being very narrowly isolated but many being widely distributed with overlapping ranges (McArthur et al., 1986). Within these overlapping subspecies they are often found growing together. Despite these morphological and ecological differences little if any work has been done on the ability of pollinators to discriminate between subspecies, resulting in a significant gap in our knowledge of how these subspecies are maintained.

Morphology


First I will read in a csv that shows morphological measurements for each genotype of rabbitbrush at each of the 5 research sites.

site color decid corolla_length corolla_width anther_length
Midvale Gray FALSE 5.00 1.00 2.50
Midvale Gray FALSE 5.00 1.25 3.00
Midvale Gray FALSE 5.00 1.00 3.00
Midvale Gray FALSE 5.50 1.00 2.00
Midvale Gray FALSE 5.50 1.00 2.00
Midvale Green FALSE 4.00 1.25 3.50
Midvale Green FALSE 4.25 1.00 2.25
Midvale Green FALSE 4.00 1.00 3.00
Midvale Green FALSE 4.50 1.25 3.00
Midvale Green FALSE 4.50 1.00 3.00
Inlet_Park Gray FALSE 5.00 1.00 5.50
Inlet_Park Gray FALSE 5.00 1.00 2.50
Inlet_Park Gray FALSE 5.00 1.00 4.00
Inlet_Park Gray FALSE 5.00 1.00 3.50
Inlet_Park Gray FALSE 4.00 1.00 3.00
Inlet_Park Green TRUE 4.00 0.75 1.50
Inlet_Park Green TRUE 3.50 1.00 1.50
Inlet_Park Green TRUE 4.00 1.00 1.50
Inlet_Park Green TRUE 3.50 0.75 1.50
Inlet_Park Green TRUE 3.00 1.00 4.00
Bluffdale Gray FALSE 5.00 1.00 3.50
Bluffdale Gray FALSE 5.00 1.00 3.00
Bluffdale Gray FALSE 5.00 1.00 4.00
Bluffdale Gray FALSE 4.50 1.00 3.00
Bluffdale Gray FALSE 4.00 1.00 3.50
Bluffdale Gray FALSE 8.00 1.25 3.00
Bluffdale Gray FALSE 6.50 1.25 3.00
Bluffdale Gray FALSE 7.00 1.00 2.50
Bluffdale Gray FALSE 7.00 1.00 3.00
Bluffdale Gray FALSE 7.50 1.25 3.25
Bluffdale Green FALSE 4.00 1.00 3.00
Bluffdale Green FALSE 4.50 1.00 3.00
Bluffdale Green FALSE 4.00 1.00 4.00
Bluffdale Green FALSE 4.50 1.00 3.50
Bluffdale Green FALSE 5.00 1.00 2.50
Provo_Mouth Gray TRUE 6.50 2.00 4.00
Provo_Mouth Gray TRUE 6.00 1.00 4.00
Provo_Mouth Gray TRUE 5.50 1.00 2.00
Provo_Mouth Gray TRUE 5.50 1.00 3.00
Provo_Mouth Gray TRUE 5.50 1.00 2.00
Provo_Mouth Gray TRUE 7.00 1.50 2.50
Provo_Mouth Gray TRUE 7.50 1.50 3.50
Provo_Mouth Gray TRUE 5.50 1.00 3.00
Provo_Mouth Gray TRUE 7.00 1.50 3.50
Provo_Mouth Gray TRUE 6.00 1.50 3.00
Provo_Mouth Green TRUE 6.00 1.00 1.50
Provo_Mouth Green TRUE 5.50 1.00 2.50
Provo_Mouth Green TRUE 5.50 1.00 3.00
Provo_Mouth Green TRUE 4.50 0.80 2.25
Provo_Mouth Green TRUE 4.50 0.80 2.00
Canyon_View Gray TRUE 7.00 1.25 3.00
Canyon_View Gray TRUE 7.00 1.00 2.00
Canyon_View Gray TRUE 6.00 1.00 2.00
Canyon_View Gray TRUE 6.00 1.50 2.50
Canyon_View Gray TRUE 6.50 1.00 2.50
Canyon_View Gray TRUE 6.00 1.00 3.00
Canyon_View Gray TRUE 6.00 1.00 3.00
Canyon_View Gray TRUE 6.00 1.00 3.00
Canyon_View Gray TRUE 7.00 1.00 4.00
Canyon_View Gray TRUE 6.50 1.50 3.50
Canyon_View Green TRUE 6.00 1.00 1.50
Canyon_View Green TRUE 5.50 1.00 2.00
Canyon_View Green TRUE 4.80 1.50 2.15




Seed Germination


One important difference in many plants is the success of their seeds. Seed germination and days till germination can vary widely between plant species. These factors may cause speciation within clusters of sympatrically growing seed plants (Willis et al., 2014). Some plants will put large energy resources into their seeds, allowing for a greater percent germination of their seeds. On the other hand some plants can create more seeds with lower germination and succeed in spreading many propagules with a small but acceptable germination rate. The timing of the germination event can also play a key role in ecological separation as a delay of several weeks may cause plants with differing germination to develop differently.

This data set shows germination percentages for each site and how many days it took each genotype to germinate.

site germ_per day_till color lat long elevation_m decid
Tabiona 50 17.00 Gray 40.35184 -110.7108 1985 TRUE
Tabiona 10 5.00 Green 40.35184 -110.7108 1985 FALSE
Tabiona 30 9.00 Gray 40.35184 -110.7108 1985 TRUE
Tabiona 50 11.80 Green 40.35184 -110.7108 1985 FALSE
Provo_Overlook 30 50.00 Gray 40.32350 -111.6326 1612 FALSE
Canyon_View 20 41.50 Gray 40.32505 -111.6441 1491 TRUE
Canyon_View 30 9.00 Gray 40.32505 -111.6441 1491 TRUE
Canyon_View 30 10.30 Green 40.32505 -111.6441 1491 TRUE
Canyon_View 50 14.00 Green 40.32505 -111.6441 1491 TRUE
Lehi_Highway 40 15.75 Green 40.44369 -111.9065 1406 TRUE
Lehi_Highway 50 12.40 Green 40.44369 -111.9065 1406 TRUE
Lehi_Highway 20 5.00 Gray 40.44369 -111.9065 1406 FALSE
Lehi_Highway 40 13.00 Gray 40.44369 -111.9065 1406 FALSE
MTEC_Lehi 70 7.57 Green 40.42186 -111.8921 1383 TRUE
MTEC_Lehi 80 6.25 Green 40.42186 -111.8921 1383 TRUE
MTEC_Lehi 10 53.00 Gray 40.42186 -111.8921 1383 TRUE
Inlet_Park 20 5.00 Gray 40.35732 -111.8982 1371 FALSE
Inlet_Park 30 4.00 Gray 40.35732 -111.8982 1371 FALSE
Inlet_Park 80 17.80 Green 40.35732 -111.8982 1371 TRUE
Inlet_Park 50 15.20 Green 40.35732 -111.8982 1371 TRUE
Bluffdale 20 10.00 Gray 40.48013 -111.9251 1363 FALSE
Bluffdale 50 25.00 Green 40.48013 -111.9251 1363 FALSE
Bluffdale 30 19.00 Green 40.48013 -111.9251 1363 FALSE
Bluffdale 40 12.00 Gray 40.48013 -111.9251 1363 FALSE
Midvale 40 5.00 Gray 40.59948 -111.9191 1306 FALSE
Midvale 50 8.00 Gray 40.59948 -111.9191 1306 FALSE
Midvale 80 15.00 Green 40.59948 -111.9191 1306 FALSE
Midvale 70 20.00 Green 40.59948 -111.9191 1306 FALSE
Provo_Mouth 10 8.00 Gray 40.30984 -111.6568 1465 TRUE
Provo_Mouth 20 15.00 Gray 40.30984 -111.6568 1465 TRUE
Provo_Mouth 70 22.00 Green 40.30984 -111.6568 1465 TRUE
Provo_Mouth 50 29.00 Green 40.30984 -111.6568 1465 TRUE



Now to looking at how morphological traits and germination rates relate.


This box plot shows us that green genotypes have a higher germination percent and shorter corolla length compared to the gray genotype that has lower germination percent and longer corolla lengths. Let’s do some statistical testing to see if any of these depictions are significant!

## List of 10
##  $ statistic  : Named num -17.7
##   ..- attr(*, "names")= chr "t"
##  $ parameter  : Named num 126
##   ..- attr(*, "names")= chr "df"
##  $ p.value    : num 5.07e-36
##  $ conf.int   : num [1:2] -35.8 -28.6
##   ..- attr(*, "conf.level")= num 0.95
##  $ estimate   : Named num [1:2] 5.38 37.62
##   ..- attr(*, "names")= chr [1:2] "mean of x" "mean of y"
##  $ null.value : Named num 0
##   ..- attr(*, "names")= chr "difference in means"
##  $ stderr     : num 1.82
##  $ alternative: chr "two.sided"
##  $ method     : chr "Welch Two Sample t-test"
##  $ data.name  : chr "Morph_Germ$corolla_length and Morph_Germ$germ_per"
##  - attr(*, "class")= chr "htest"

Running a Ttest helps to see that since the p-value is < 0.5 that the germination percent and corolla length depending on the genotype of rabbitbrush is significant. There is quite a difference between the two color groups.

Flowering Phenology


Flowering phenology and germination go hand in hand. The start and end flowering time of each genotype at each site was observed and that is what the below data set is depicting.
Cleaning up the phenology data set.
site color flower_start flower_end lat long elevation_m
Midvale Gray 2022-09-06 18:00:00 2022-10-19 18:00:00 40.59948 -111.9191 1306
Midvale Green 2022-08-29 18:00:00 2022-10-05 18:00:00 40.59948 -111.9191 1306
Inlet_Park Gray 2022-08-29 18:00:00 2022-10-10 18:00:00 40.35732 -111.8982 1371
Inlet_Park Green 2022-08-29 18:00:00 2022-10-01 18:00:00 40.35732 -111.8982 1371
Bluffdale Gray 2022-09-12 18:00:00 2022-10-19 18:00:00 40.48013 -111.9251 1363
Bluffdale Green 2022-09-14 18:00:00 2022-10-23 18:00:00 40.48013 -111.9251 1363
Provo_Mouth Gray 2022-10-03 18:00:00 2022-10-23 18:00:00 40.30984 -111.6568 1465
Provo_Mouth Green 2022-10-12 18:00:00 2022-10-23 18:00:00 40.30984 -111.6568 1465
Canyon_View Gray 2022-09-14 18:00:00 2022-10-14 18:00:00 40.32505 -111.6441 1491
Canyon_View Green 2022-09-21 18:00:00 2022-10-23 18:00:00 40.32505 -111.6441 1491

Messing around with the flowering start and end time for each color group.

## Rows: 10
## Columns: 2
## Groups: color [2]
## $ color            <chr> "Gray", "Gray", "Gray", "Gray", "Gray", "Green", "Gre…
## $ mean_flower_time <drtn> 43 days, 42 days, 37 days, 20 days, 30 days, 37 days…

## List of 10
##  $ statistic  : Named num 10.2
##   ..- attr(*, "names")= chr "t"
##  $ parameter  : Named num 9
##   ..- attr(*, "names")= chr "df"
##  $ p.value    : num 3.04e-06
##  $ conf.int   : num [1:2] 25.2 39.6
##   ..- attr(*, "conf.level")= num 0.95
##  $ estimate   : Named num 32.4
##   ..- attr(*, "names")= chr "mean of x"
##  $ null.value : Named num 0
##   ..- attr(*, "names")= chr "mean"
##  $ stderr     : num 3.18
##  $ alternative: chr "two.sided"
##  $ method     : chr "One Sample t-test"
##  $ data.name  : chr "Phenology$flower_time"
##  - attr(*, "class")= chr "htest"
##             Df Sum Sq Mean Sq F value Pr(>F)  
## color        1   40.0   40.00    2.54  0.186  
## site         4  805.4  201.35   12.78  0.015 *
## Residuals    4   63.0   15.75                 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1




Pollen Flow With UV Pigments as a Proxy for Pollen


Pollen flow is such an important factor to consider when studying plant and pollinator interactions.
Loading and cleaning the pollen data set


Loading and cleaning another pollen data set with all the pigment found at each site for a duration of 40 minutes per site.

Pollinators